RFC: PATCH for thread-safe C++ static local initialization
Sun Aug 22 16:44:00 GMT 2004
email@example.com (Jason Merrill) wrote on 20.08.04 in <firstname.lastname@example.org>:
> However, the current gthr interface doesn't actually support recursive
> mutexes; I don't think appropriate APIs exist on all targets. It's
> possible to emulate a recursive mutex, as I do in this patch, but my
> emulation relies on being able to request a unique identifier for the
> current thread, which is also not part of the gthr interface--though it is
> part of the objc thread interface.
Are there good reasons not to unify those interfaces? They certainly look
as if they started by one being cloned from the other anyway.
More information about the Gcc-patches